Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim):

A New Discrete-event Simulator

Hanene Rouainia

1 a

, Hanen Grichi

2,3 b

, Laid Kahloul

4 c

and Mohamed Khalgui

3,5,∗ d

Faculty of Sciences of Tunis, El-Manar University, Tunis, Tunisia

Faculty of Sciences of Bizerte (FSB), University of Carthage, Bizerte, Tunisia

School of Electrical and Information Engineering, Jinan University, Zhuhai, China

LINFI Laboratory, Computer Science Department, Biskra University, Biskra, Algeria

INSAT Institute, University of Carthage, Tunis, Tunisia

Keywords:

Wireless Sensor Network, Reconﬁgurable Wireless Sensor Network, RWSNSim, Simulator, Sensor Node,

Sink Node, Agent.

Abstract:

Reconﬁgurable wireless sensor networks become an important area in research and industrial communities.

With their development and spread, many problems and solutions have emerged. Network simulators have

become an essential necessity to study the impact of these solutions on networks in order to avoid huge costs

in terms of money, time, and effort if applied on the ground. In this paper, we propose a new discrete-

event simulator for WSNs and RWSNs called RWSNSim. We present its description, modeling, and provided

services. Finally, to demonstrate the efﬁciency of RWSNSim, we simulate a case study and detail the simulator

functioning steps.

1 INTRODUCTION

With the spread of microelectromechanical sys-

tems (MEMS) technology, wireless sensor networks

(WSNs) have gained worldwide attention. WSNs de-

ploy a set of multi-functional devices known as sensor

nodes (SNs). They have considerable characteristics

such as small size, low cost and computing resources,

and limited processing. Sensor nodes can sense both

physical and chemical measurements in the surround-

ing environment to process the sensing data, com-

municate with each other wirelessly, and work co-

operatively (Agrawal, 2017), (Rouainia et al., 2022).

WSNs can be used in a variety of areas like medical,

environmental monitoring, military, and smart homes

(Khriji et al., 2018), (Vijayalakshmi and Muruganand,

2018).

We have several challenges in WSNs like lack

of energy problem which occurs because the WSNs

work under many types of renewable energy re-

sources which are not frequently available, real-time

https://orcid.org/0000-0001-7544-988X

https://orcid.org/0000-0002-4601-3574

https://orcid.org/0000-0002-9739-7715

https://orcid.org/0000-0001-6311-3588

∗

Member IEEE

problem which means that in some applications, the

transmitted messages over the network must respect

their deadlines, and packet dropping which may be

caused by software or hardware failures as men-

tionned by several researchers in literature (Rouainia

et al., 2020), (Rouainia et al., 2022), (Haﬁdi et al.,

2020).

Reconﬁgurable wireless sensor networks

(RWSNs) are wireless sensor networks with the

possibility to execute reconﬁguration scenarios such

as mobility and resizing. The existence of additional

speciﬁc devices (i.e., mobile sensor nodes, mobile

sink nodes, and software and hardware agents) allows

RWSNs to execute reconﬁguration scenarios. These

reconﬁguration scenarios are proposed as solutions to

the mentioned WSNs problems in our previous works

(Rouainia et al., 2020), (Rouainia et al., 2022).

Since the expensive cost, effort, time, and com-

plexity implicated in the construction and the imple-

mentation of RWSNs, the developers prefer to get an

overview about feasibility and behaviour of RWSNs

before hardware implementation. Indeed, the anal-

ysis and evaluation of the proposed solutions and

techniques through real experiments are not feasible,

complex, and very expensive in terms of time, effort,

and cost. As result, to keep up with these challenges,

Rouainia, H., Grichi, H., Kahloul, L. and Khalgui, M.

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator.

DOI: 10.5220/0011318300003266

In Proceedings of the 17th International Conference on Software Technologies (ICSOFT 2022), pages 349-361

ISBN: 978-989-758-588-3; ISSN: 2184-2833

349

several simulation tools (Nayyar and Singh, 2015) are

proposed to test and analyse the performance of the

proposed techniques, protocols, and solutions. They

have many advantages like low cost, easy develop-

ment, giving real-time results, and detecting the posi-

tive and negative effects on the entire network.

There are two types of simulations surround-

ing WSNs and RWSNs: trace-driven simulation and

discrete-event simulation. Trace-driven simulation is

an important approach in many simulation applica-

tions, especially in real-time applications. It enables

fast design evaluation by considering system models

which are derived from a sequence of observations

made on a real system. It allows users to get in-depth

details of the simulation model. But it contains sev-

eral drawbacks like increasing the complexity of the

simulation. On the other hand, discrete-event simu-

lation is used to model real-world systems that can

be divided into several logically separate processes

that autonomously progress through time. This type

of simulation is mostly used in WSNs and RWSNs

due to its ease in simulating various tasks running on

different sensor nodes, sinks, and agents (Nayyar and

Singh, 2015), (Rouainia et al., 2022).

We can evaluate and compare the simulators of

WSNs and RWSNs using a set of parameters includ-

ing the following: the type of simulator which is clas-

siﬁed into three categories: generic, code level, and

ﬁrmware simulator, the license which can be commer-

cial or open-source, the platform which is the operat-

ing system on which the simulator operates such as

Windows, Linux, or both, and WSN platforms which

are deﬁned in terms of sensors types and platforms

which can be simulated by the simulator (Nayyar and

Singh, 2015).

In this paper, we propose a new simulator

RWSNSim to construct WSNs and RWSNs, save them

in a database, use two routing protocol (LEACH

and WBM-TEEN), plot the simulation graph, present

an execution report for each monitoring time, draw

the resulting line charts after the simulation, and

compare between the different networks and simu-

lations (RWSNSim, 2022). It permits also to apply

the methodology proposed in (Rouainia et al., 2022)

which is a new energy efﬁcient and fault tolerant

methodology based on a multi-agent architecture in

RWSNs using mobile sink nodes, mobility, resizing,

and test packet technique. The proposed simulator is

considered as a discrete-event simulator.

The rest of the paper is organized as follows. Sec-

tion 2 presents the related works. The new simulator

is developed in Section 3. Section 4 exposes a case

study executed by RWSNSim. Finally, the conclusion

is drawn in Section 5.

2 RELATED WORKS

Several WSNs simulation tools have been proposed

by academic and commercial communities (Nayyar

and Singh, 2015), (Rajan et al., 2015). In this section,

we will discuss some of the most important work in

this ﬁeld.

RWiN-Environment (RWiN, 2016) is a graphical

tool developed to evaluate the services of the RWiN-

Methodology which is proposed to analyse, construct,

develop, and verify an RWSN system in order to re-

duce the consumed energy by the network (Grichi

et al., 2016a), (Grichi et al., 2016b).

Network Simulator-3 (NS-3) (NS3, 2022) is a

discrete-event simulator for Internet systems. It was

launched in June 2008 as an open-source project.

It is free, targeted primarily at research and educa-

tional uses, licensed under the GNU GPLv2 license,

and maintained by a worldwide community. Net-

work Simulator-3 is written in C++ language and

python. The simulations executed in NS-3 can be im-

plemented using pure C++ with optional python bind-

ings. It can work in various operating system plat-

forms like Linux and Windows via Cygwin. The lat-

est version of NS-3 is NS-3.35 which is released in

October 1, 2021. It provides several improvements

compared to previous versions like IPv6 support for

NixVectorRouting and a group mobility helper.

OMNet++ (OMNeTPP, 2019) is a powerful

object-oriented discrete-event simulator. It can be

used for the simulation of computer networks, dis-

tributed and parallel systems, modeling of multipro-

cessors, and performance evaluation of complex soft-

ware systems. It was launched in September 1997

and has a large number of users in academic, ed-

ucational, and research-oriented commercial institu-

tions. Indeed, OMNeT++ is not a simulator, but

it permits writing simulation scenarios using several

frameworks and tools. It is considered as an extensi-

ble, modular, and component-based open-architecture

simulation framework implemented using C++ pro-

gramming language. It provides an extensive graph-

ical user interface (GUI) and intelligence support.

OMNET++ distributions are available for different

operating systems like Windows, Linux, and MAC

OS X. The latest version of OMNeT++ is 5.7 which

is released in October 6, 2021 and is intended to be

the last release of the 5.x series.

JavaSim (J-Sim) simulator (JSim, 2022) is

an object-oriented simulation package based upon

C++SIM started in year 1997. It is considered as a

general purpose simulator used by many commercial

and academic organizations. J-Sim is a platform open

source, free, extensible, neutral, and reusable because

ICSOFT 2022 - 17th International Conference on Software Technologies

350

Table 1: Comparison of some simulation tools.

Name Type Program. License GUI Scalability Portability Designed for

Language (WSNs or

RWSNs)

RWiN- Discrete Java Open Excellent Medium Yes RWSNs

Environment event source

NS-3 Discrete C++, Open Limited Large Yes WSNs

event Python source

OMNet++ Discrete C++ Open Excellent Large Yes WSNs

event source

JavaSim Discrete Java Open Medium Medium Yes WSNs

event source

GloMoSiM Discrete C Open Limited Large Yes WSNs

event source

RWSNSim Discrete Java Open Excellent Medium Yes WSNs, RWSNs

event source

it has been developed using Java, based on ACA

(Autonomous Component Architecture), and offers

discrete-event process-based simulation. The use of

a script interface in J-Sim permits executing several

script-based languages such as Python and Perl. The

current version of JavaSim is 2.2.0 GA which is re-

leased in January 4, 2020. J-Sim offers a framework

for WSN simulation using INET, ACA, and Wireless

Protocol stack.

Global Mobile Information System Simulator

(GloMoSiM) (GloMoSim, 2020) is a discrete-event

scalable simulation software used for large-scale

wireless and wired network systems and developed

by Parallel Computing Lab at UCLA. It is designed

solely for wireless networks using parallel discrete-

event simulation provided by Parsec. It compiles the

simulation of protocols using Parsec compiler. Glo-

MoSiM supports three communication protocols: tra-

ditional Internet protocols, multi-hop wireless com-

munication, and direct satellite communication. It

permits to simulate networks with thousands of het-

erogeneous nodes and communication links.

Finally, the proposed simulator RWSNSim

(RWSNSim, 2022) is considered as the ﬁrst simula-

tion tool designed for both WSNs and RWSNs. Table

1 presents a comparison of the mentioned simulation

tools according to general parameters.

Through table 1, we conclude that the strengths of

RWSNSim are summarized in the following: license,

graphical user interface (GUI), portability and its abil-

ity to simulate both WSNs and RWSNs.

3 RWSNSim: RECONFIGURABLE

WIRELESS SENSOR

NETWORKS SIMULATOR

In this section, we describe the RWSNSim simulator

and present its provided services. We also model the

architecture of WSNs and RWSNs using class dia-

grams and validate them. Furthermore, we model the

used databases to store the information of WSNs and

RWSNs using entity relationship diagrams.

3.1 Description

RWSNSim is a descrete-event network simulator writ-

ten in Java. It is developed using Java for wireless

sensor networks and reconﬁgurable wireless sensor

networks. It is an open source, free, reusable and

extensible simulator. It is designed for research and

educational use. RWSNSim is capable of simulating

networks that contain hundreds of sensor nodes, sink

nodes, and zone agents. It supports several operating

systems such as Windows and UNIX.

The proposed simulator is a desktop simulator that

makes it possible to construct WSNs and RWSNs and

simulate them using two routing protocols (LEACH

and WBM-TEEN). It can also apply the proposed

methodology in (Rouainia et al., 2022) which is a

new energy efﬁcient and fault tolerant methodology

in RWSNs. This methodology based on a multi-agent

architecture and consists of the implementation of a

set of techniques such as the use of mobile sink nodes,

mobility, resizing of zones, and test packet technique.

RWSNSim provides several services during the

simulation process, including the following:

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator

351

• Creating WSNs and RWSNs and saving them in a

database using Java, MySQL or hsqldb according

to the user choice.

• Providing two routing protocols: LEACH and

WBM-TEEN which are considered as energy ef-

ﬁcient protocols.

• Executing the simulation graph of each network

using jgraph, jgraphx, and jgrapht libraries.

• Displaying the execution report for each monitor-

ing time.

• Extracting the simulation results and represent-

ing them in form of line charts using jfreechart

library.

• Comparing the different networks and simula-

tions.

3.2 WSNs and RWSNs Modeling

In the following, we describe the target sensor net-

works used by RWSNSim simulator. Then, we present

the architecture of WSNs and RWSNs with their com-

ponents and relations between them using class dia-

grams which are used to develop the RWSNSim sim-

ulator. We also validate the used class diagrams by

a natural language and with ROCL constraints. This

last one is an extension of OCL which improves the

speciﬁcation and validation of constraints related to

different execution scenarios of systems (Grichi et al.,

2015).

3.2.1 The Target Sensor Networks

(a) Wireless Sensor Networks (WSNs):

With RWSNSim, each wireless sensor network

composed of:

• A base station BS plays the role of a gateway

between the WSN and its administrator. It re-

ceives the sensing data from the gateways and

analyses them, which allows the administrator

of the network to take an overview of the net-

work and to call human intervention in some

cases.

• A set of zones.

• Each zone contains a gateway that collects the

sensing data from the cluster heads and sends

them to the base station.

• Each zone contains a set of clusters.

• Each cluster contains a cluster head that col-

lects the sensing data from its cluster sensor

nodes and sends them to its zone gateway.

• Each cluster contains a set of sensor nodes that

monitor the chemical and physical measure-

ments in the surrounding environment and send

them to their cluster head.

(b) Reconﬁgurable Wireless Sensor Networks

(RWSNs):

With RWSNSim, each reconﬁgurable wireless sen-

sor network composed of:

• A base station BS plays the role of a gateway

between the RWSN and its administrator. It

receives alert messages from the zone agents,

analyses them and communicates with the ad-

ministrator of the network to call human inter-

vention in some cases.

• A controller agent Ag

Ctrl

controls the whole

network by executing several tasks such as re-

ceiving the sensing data from the zone agents,

analysing them, sending alert messages to the

BS, and applying the resizing of zones.

• A set of zones.

• Each zone contains a zone agent which controls

the zone by executing several tasks such as or-

ganizing the sensor nodes in subzones into clus-

ters, applying the mobility, and detecting the

malfunctioning sink nodes.

• Each zone contains a set of subzones.

• Each subzone contains a mobile sink node that

executes several tasks like collecting the sens-

ing data from the cluster heads in the subzone,

sending them to the zone agent, and detecting

the malfunctioning sensor nodes. Each mobile

sink node is controlled by a software agent in-

stalled in the sink itself named sink agent.

• Each subzone contains a set of clusters.

• Each cluster contains a cluster head that col-

lects the sensing data from the sensor nodes in

the cluster and sends them to the sink node.

• Each cluster contains a set of sensor nodes that

monitor the chemical and physical measure-

ments in the surrounding environment and send

them to their cluster head. Each sensor node is

controlled by a software agent installed in the

sensor node itself named node agent.

3.2.2 Class Diagrams

Figure 1 presents the class diagram of wireless sensor

networks which is used to develop RWSNSim. In this

class diagram, we have 11 classes of which we deﬁne

the following:

• WSN: It models a WSN and contains the differ-

ent thresholds, constants, and consumption val-

ues. They can be used by the different entities

ICSOFT 2022 - 17th International Conference on Software Technologies

352

Figure 1: WSN class diagram.

in WSNs such as an energy threshold [Beta], and

vectors of consumed values of energy and time

parameters [Times, TimeSens, Cons, ConsSens].

• BaseStation: it models the base station BS. It con-

tains the BS principal properties such as its po-

sition [position], vector of zones [VZ] and oper-

ations like organizing the nodes in each zone Z

[organizeNodes()], sending an order to a gateway

[sendOrderTo(G: Gateway)].

• Zone: it models each zone Z

and contains its

principal properties such as an identiﬁant [IDk],

its state [state], and a vector of clusters [VCls].

• Gateway: it models each gateway G

and contains

its principal properties like an identiﬁant [IDk]

and a name [Name] and operations such as receiv-

ing an order from BS to collect the sensing data

[receptOrderFrom(B: BaseStation)] and sending

a vector of sensing data to BS [sendVSDTo(B:

BaseStation)].

• Cluster: it models each cluster in Z

. It contains

the clusters’ principal properties like identiﬁants

[IDs] and a vector of sensor nodes belonging to

this cluster [VN].

• Node: it models each sensor node N

i,k

in WSN. It

extends from the Entity class and contains the sen-

sor nodes’ principal properties such as identiﬁants

[IDs], a vector of batteries charges [Charges], and

a vector of sensors [VS] and operations like mon-

itoring physical or chemical measurements in the

environment [monitoring(): SensData[NbS]] and

sending a vector of sensing data to the successor

[sendVSDTo(VSD: SensData, E: Entity)].

Figure 2 presents the class diagram of reconﬁgurable

wireless sensor networks which is used to develop

RWSNSim. In this class diagram, we have 17 classes

of which we deﬁne the following:

• RWSN: it models a reconﬁgurable wireless sen-

sor network. It contains the different constants,

variables, energy and time consumption values.

They can be used by the different entities in

RWSNs such as (number of zones [NbZ] and vec-

tors of thresholds and consumed values of energy

and time parameters [Thresholds, SensThr, Cons,

ConsSens]).

• Entity: is the class model of all physical entities in

RWSNs. It contains their common properties like

the state of the entity [state] and its position [po-

sition] and operations such as activating the entity

[activate()] and deactivating it [deactivate()].

• AgCtrl: it models the controller agent Ag

Ctrl

. It

extends from the Entity class and contains the

Ctrl

principal properties like a vector of zones

[VZ] and a vector of sensing data [VSD] and op-

erations such as applying the resizing task [re-

sizing(Za: Zone, Zb: Zone)], sending an order

to a zone agent [sendOrderTo(Ag: ZoneAgent)],

sending a test packet to zone agents to detect

the malfunctioning zone agent [sendTstPckTo(Ag:

ZoneAgent)], and isolate a malfunctioning zone

agent [isolate(Ag: ZoneAgent)].

• ZoneAgent: it models each zone agent Ag

RWSN. It extends from the Entity class and con-

tains the zone agents’ principal properties like an

identiﬁant [IDk], a vector of sensor nodes [VN],

and a vector of subzones [VSZ] and operations

such as organizing the sensor nodes in Z

[orga-

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator

353

Figure 2: RWSN class diagram.

nizeNodes(): SubZone[*]], applying the mobil-

ity task [applyMobility(E: Entity)], and isolate the

malfunctioning sinks and sensor nodes [isolate(E:

Entity)].

• Sink: it models each mobile sink node SN

m,k

in RWSN. It extends from the Entity class and

contains the mobile sink nodes’ properties like

identiﬁants [IDs], new position coordinations

[newposition], and a boolean property [free]

and operations such as moving to another po-

sition [moveTo(position2: Integer[3])], sending

an acknowledge message to the zone agent Ag

[sendAckMsgTo(Ag: ZoneAgent)] and sending a

test packet to a sensor node N

i,k

[sendTstPckTo(N:

Node)].

• Sensor: it models each sensor Sens

j,N

i,k

in a sen-

sor node N

i,k

. It extends from the Entity class and

contains the sensors’ principal properties such as

a sensing value [SD].

3.2.3 Validation Model

In order to validate the class diagrams which are pre-

sented in ﬁgure 1 and 2, we use natural language and

ROCL. The natural language permits to describe the

different constraints that must be respected by the dif-

ferent entities in each network. While, ROCL allows

formalizing them. We present in the following some

constraints that must be respected by some entities in

an RWSN.

A. Natural Language - Constraint 01

• AgCtrl Class: in this class, we have two impor-

tant constraints which are explained as follows:

- Before the application of the resizing task be-

tween two zones Z

and Z

, the following condi-

tions must be met:

a- The number of active nodes in Z

must be less

than or equal to Λ = T hresholds[3] .

b- Z

must be a neighbor zone of Z

which con-

tains the minimum number of active nodes.

- After the application of the resizing task between

and Z

the following results must be met:

a- The number of zones will be decreased.

b- The number of nodes in Z

will be increased.

c- The zone agent Ag

and Z

will be deactivated.

d- The number of sinks in Z

will be increased.

e- The number of Z

neighbors will be increased.

B. Natural Language - Constraint 02

• ZoneAgent Class: in this class, we have two im-

portant constraints which are deﬁned as follows:

- If the total charge of a node N

i,k

∈ S

is less than

or equal to α = T hresholds[1], the zone agent Ag

must apply the mobility task.

- After the application of the mobility, the deac-

tivation or the isolation of sensor nodes or sink

nodes in a zone Z

, Ag

must organize the order

of nodes in Z

in subzones into clusters.

ICSOFT 2022 - 17th International Conference on Software Technologies

354

ROCL - Constraint 01

context AgCtrl, RW SN, ZoneAgent Aga INV

resizing(Zone Za, Zone Zb) :: Execute()

IMPLIESdeacivate(Aga)ANDdeactivate(Za)

pre : Za.NbNAct ≤ T hresholds[3] AND

Za.V Neigh → includes(Zb) AND Zb.NbNAct ≤

Za.V Neigh.allInstances → NbNAct

post : NbZ = NbZ − 1 AND

Zb.NbN = Zb.NbN + Za.NbN

deacivate(Za) :: Execute() IMPLIES

Zb.NbSink = Zb.NbSink + Za.NbSinkAND

Zb.NbNeigh = (Zb.NbNeigh − 1) + Za.NbNeigh

AND Zb.V Neigh = Zb.V Neigh −

Za → union(Za.V Neigh)

ROCL - Constraint 02

context ZoneAgent Ag, Entity E,

NodeN, SinkSN, MNodeMNINV

IF (N.charges[2] ≤ α) IMPLIES

(applyMobility(MN) OR applyMobility(SN))

context deactivate(Entity E) :: Execute()

(Node, Sink) → includes(sel f ) IMPLIES

organizeNodes()

context Node N, ZoneAgent Ag, Entity E,

Ag.applyMobility(Entity E) :: Execute()

(MNode, Sink) → includes(sel f ) IMPLIES

organizeNodes()

C. Natural Language - Constraint 03

• Sink Class and MNode Class: in these classes,

we have an important constraint which is deﬁned

as follows:

- When a zone agent Ag

applies the mobility of a

mobile entity E where E is a mobile sink node or

a mobile node. The following changes will take

place:

a- The position of E must be changed.

b- The charge of the principal battery of E must

be decreased.

ROCL - Constraint 03

context Sink E, MNodeE, E.moveTo(position2 :

Integer[3]) :: Execute()

post :E.Charges[0] = E.charges[0]−

Emob(newposition, position2)

AND E.newposition = position2

3.3 Databases Modeling

To create the databases used to store the information

related to the simulated WSNs and RWSNs, we model

them using entity relationship diagrams (ERD). In-

deed, we have two ERDs:

Figure 3: Entity relationship diagram for WSNs used by

RWSNSim.

1) ERD for WSNs: Figure 3 presents the entity

relationship diagram of wireless sensor networks used

in RWSNSim to generate the databases used to store

the wireless sensor networks’ information entered by

the user or generated automatically. It contains the

following databases:

• [DB: WSNs]: is a database that includes a sin-

gle entity named WSN Infos. It is a weak entity

which contains the general information of each

wireless sensor network such as the WSN name

[WSN Name] which is the key attribute, the num-

ber of zones [NbZones], and the used protocol

[protocol].

• [DB: WSN Name]: For each WSN created by the

user, a database with the same name is created to

store its information. It includes three entities de-

ﬁned as follows:

– Zone: is a strong entity which contains the gen-

eral information of each zone like an identiﬁ-

ant [IDk] which is the key attribute, its state

[State], and the number of sensor nodes it con-

tains [NbNodes].

– Gateway: is a strong entity which contains the

general information of each gateway in the net-

work such as its identiﬁant [IDk] which is the

key attribute, its coordinates which presents its

position in the network environment [Coordi-

nates], and its state [State].

– Node: is a strong entity which contains the gen-

eral information of each sensor node in the net-

work such as two identiﬁants [IDi, IDk] which

are the key attributes, the initial charge of the

principal battery [Cpr], and the initial charge

of the additional battery [Cad].

2) ERD for RWSNs: Figure 4 presents the en-

tity relationship diagram of reconﬁgurable wireless

sensor networks used in RWSNSim to generate the

databases used to store the reconﬁgurable wireless

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator

355

Figure 4: Entity relationship diagram for RWSNs used by RWSNSim.

sensor networks information entered by the user or

generated automatically. It contains the following

databases:

• [DB: RWSNs]: is a database that includes a sin-

gle entity named RWSN Infos. It is a weak entity

which contains the general information of each re-

conﬁgurable wireless sensor network such as the

RWSN name [RWSN Name] which is the key at-

tribute, the number of agents whatever their types

[NbAgents], and the number of sink nodes [Nb-

Sinks].

• [DB: RWSN Name]: For each RWSN created by

the user, a database with the same name is created

to store its information. It includes four entities

deﬁned as follows:

– Zone: is a strong entity which contains the gen-

eral information about each zone like an identi-

ﬁant [IDk] which is the key attribute, its name

[Name], and the list of neighbor zones [Neigh-

bors].

– Agent: is a strong entity which includes the

general information about each zone agent in

the network such as its name [Name] which is

the key attribute, its coordinates which present

its position in the network environment [Coor-

dinates], and its state [State].

– Sink: is a strong entity which contains the gen-

eral information of each sink node in the net-

work like two identiﬁants [IDm, IDk] which are

the key attributes, its type which can be ’ﬁxed’

or ’mobile’ [Type], and the total charge of its

batteries [Ctot].

– Node: is a strong entity which includes the gen-

eral information of each sensor node in the net-

work such as two identiﬁants [IDi, IDk] which

are the key attributes, its coordinates which

present its position in the network environment

[Coordinates], and its type which can be ’ﬁxed’

or ’mobile’ [Type].

4 CASE STUDY

In this case study, we simulate a reconﬁgurable wire-

less sensor network to show all provided services by

RWSNSim which cannot be shown if we simulate a

wireless sensor network such as mobility and resizing

notiﬁcation windows.

We propose a reconﬁgurable wireless sensor

network denoted by RWSN UnderWater as a simu-

lation example with RWSNSim. RWSN UnderWater

is composed of a base station termed by BS,

a controller agent denoted by Ag

Ctrl

, and two

zones {Z

and Z

}. Each zone Z

|k ∈ {1, 2}

contains one zone agent Ag

, 20 sensor nodes

(12 ﬁxed nodes and 8 mobile nodes) denoted by

= {N

i,k

|i ∈ [1..20] k ∈ 1, 2}, and 3 mobile sink

nodes termed by S

= {SN

m,k

|m ∈ [1..3]}. Each sen-

sor node has two sensors: CO

sensor and CH

sensor.

A. Choosing the Appropriate Routing Protocol

In RWSNSim, we have two routing protocols:

LEACH and WBM-TEEN. Before the construction

of a WSN or an RWSN network, the user has to

choose one of them. In this case study, we chose

LEACH protocol to simulate RWSN UnderWater

network.

ICSOFT 2022 - 17th International Conference on Software Technologies

356

Figure 5: The two manners to construct RWSN UnderWater using RWSNSim.

B. Construction of RWSN UnderWater using

RWSNSim

With RWSNSim, the user can construct a WSN or an

RWSN and save it in a database. We have two man-

ners of network construction:

1) Manual and regular: depending on this manner,

all information related to each entity in the network

must be entered manually such as the coordinates and

the state of each entity.

2) Automatic and random: depending on this man-

ner, all information related to each entity in the net-

work is randomly selected by the system according

to some inputs such as the total number of zones and

the number of sensor nodes and sink nodes in each

zone. All information can be modiﬁed by the user in

the created database.

Figure 5 shows these two manners in the

RWSNSim graphic user interface.

In the proposed case study, we chose the manual

and regular manner. Therefore, we have added two

zones and the entities they contain (zone agents, mo-

bile sink nodes, and sensor nodes) with their informa-

tion.

After the construction of RWSN UnderWater net-

work, RWSNSim saves all information entered by the

user or generated automatically in a database. Figure

6 shows the RWSN UnderWater information stored in

a database using RWSNSim.

C. Execution of RWSN UnderWater using

RWSNSim

After the construction of RWSN UnderWater net-

work, the user can start its execution. Before start-

ing the network execution, RWSNSim demands a set

of parameters and thresholds which showed in ﬁg-

ure 2 and detailed in the methodology proposed in

(Rouainia et al., 2022) like α (which is an energy

charge threshold in sensor nodes used by the zone

agent Ag

to apply the mobility), γ (which is a num-

ber of active sensor nodes threshold in a zone Z

used

by the zone agent Ag

to apply the mobility), and λ

(which is a number of active sensor nodes threshold

in a zone Z

used by the controller agent Ag

Ctrl

to ap-

ply the resizing of zones). After that, two windows

appear: the simulation graph window and the execu-

tion report window.

1) Simulation graph window: it displays the dif-

ferent RWSN UnderWater network entities and their

distribution in the network and shows the communi-

cation between them in real time. Figure 7 presents

the simulation graph captured during the simulation

of RWSN UnderWater network.

2) Execution report window: It is a window

containing the different sensing data values received

by the controller agent from zone agents. Figure 8

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator

357

Figure 6: RWSN UnderWater information stored in a database using RWSNSim.

displays the execution report of RWSN UnderWater

network using RWSNSim.

D. Notiﬁcations & Alerts

During the simulation, many notiﬁcation and alert

windows appear synchronized with various events

that occur during the execution of RWSN UnderWater

network. There are four types of notiﬁcation and alert

windows deﬁned as follows:

1) Deactivation notiﬁcation window: It appears

when an entity or a zone in the network is deactivated.

Figure 9 shows the deactivation notiﬁcation window

which appears when the node N

6,1

is deactivated.

2) Mobility notiﬁcation window: It appears when

a mobile entity in the network has moved to another

position. Figure 9 illustrates the mobility notiﬁca-

tion window which appears when the mobile sink

nodes SN

1,2

and SN

2,2

have moved to other positions

in RWSN UnderWater network.

3) Resizing notiﬁcation window: It appears when

the controller agent Ag

Ctrl

applies the resizing of

zones between two zones Z

and Z

where a ∈

[1..NbZ] and b ∈ [1..NbZ]. Figure 9 illustrates the re-

sizing notiﬁcation window which appears when the

controller agent Ag

Ctrl

applies the resizing between

and Z

4) Alert window: It appears with an alert sound

when the controller agent Ag

Ctrl

received sensing

data with values exceeding the sensing data thresh-

olds which are illustrated in ﬁgure 2 and detailed

in the methodology proposed in (Rouainia et al.,

2022). Figure 9 shows the alert window which

appears when the controller agent Ag

Ctrl

received

several sensing data with values exceeding the

sensing data thresholds (CO2T hreshold = 5% and

CH4T hreshold = 10%).

E. Draw the Obtained Line Chart of

RWSN UnderWater using RWSNSim

After the simulation, RWSNSim provides the obtained

results in the form of line charts. Figure 10 displays

the obtained line chart of RWSN UnderWater.

Finally, we note that RWSNSim is not a speciﬁc

simulator developed to implement a speciﬁc method-

ology, but rather tends to be more generalized. In-

deed, it can simulate WSNs and RWSNs without ap-

plying the proposed methodology in (Rouainia et al.,

2022). It has also the scability to include new solu-

tions, techniques and routing protocols.

ICSOFT 2022 - 17th International Conference on Software Technologies

358

Figure 7: RWSN UnderWater simulation graph using RWSNSim.

Figure 8: RWSN UnderWater execution report using RWSNSim.

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator

359

Figure 9: Notiﬁcations and alert windows.

Figure 10: The obtained line chart of RWSN UnderWater.

ICSOFT 2022 - 17th International Conference on Software Technologies

360

5 CONCLUSION

In this paper, we proposed a new discrete-event

simulator designed for WSNs and RWSNs named

RWSNSim. We described the provided services by the

proposed simulator. We also modeled the architecture

of WSNs and RWSNs used by RWSNSim. Further-

more, we modeled the databases used by RWSNSim.

The proposed simulator permits simulating WSNs

and RWSNs with hundreds of entities. In the future,

we will strive to improve the proposed simulator by

adding new routing protocols, new solutions and tech-

niques, new features, and new types of charts.

REFERENCES

Agrawal, D. P. (2017). Sensor nodes (sns), camera sensor

nodes (c-sns), and remote sensor nodes (rsns). In Em-

bedded Sensor Systems, pages 181–194. Springer.

GloMoSim (2020). GloMoSim Simulator Projects

: Online Network Simulator : Network Simu-

lation Tools. https://networksimulationtools.com/

glomosim-simulator-projects/. [Online; accessed 27-

March-2022].

Grichi, H., Mosbahi, O., and Khalgui, M. (2015). Rocl:

New extensions to ocl for useful veriﬁcation of ﬂexi-

ble software systems. pages 45–52.

Grichi, H., Mosbahi, O., and Khalgui, M. (2016a). A devel-

opment tool chain for reconﬁgurable wsns. In Fujita,

H. and Papadopoulos, G. A., editors, New Trends in

Software Methodologies, Tools and Techniques - Pro-

ceedings of the Fifteenth SoMeT 16, Larnaca, Cyprus,

12-14 September 2016, volume 286 of Frontiers in Ar-

tiﬁcial Intelligence and Applications, pages 101–114.

IOS Press.

Grichi, H., Mosbahi, O., Khalgui, M., and Li, Z. (2016b).

Rwin: New methodology for the development of re-

conﬁgurable wsn. IEEE Transactions on Automation

Science and Engineering, PP:1–17.

Haﬁdi, Y., Kahloul, L., Khalgui, M., Li, Z., Alnowibet, K.,

and Qu, T. (2020). On methodology for the veriﬁca-

tion of reconﬁgurable timed net condition/event sys-

tems. IEEE Transactions on Systems, Man, and Cy-

bernetics: Systems, 50(10):3577–3591.

JSim (2022). GitHub - nmcl/JavaSim: JavaSim simula-

tion classes and examples. https://github.com/nmcl/

JavaSim. [Online; accessed 27-March-2022].

Khriji, S., Houssaini, D., Kammoun, I., and Kanoun, O.

(2018). Energy-efﬁcient techniques in wireless sensor

networks: Technology, Components and System De-

sign, pages 287–304.

Nayyar, A. and Singh, R. (2015). A comprehensive re-

view of simulation tools for wireless sensor networks

(wsns). pages 19–47.

NS3 (2022). ns-3 : a discrete-event network simulator for

internet systems. https://www.nsnam.org/. [Online;

accessed 26-March-2022].

OMNeTPP (2019). OMNeT++ Discrete Event Simulator.

https://omnetpp.org/. [Online; accessed 27-March-

2022].

Rajan, C., Geetha, K., Priya, C. R., Geetha, S., and

Manikandan, A. (2015). A simple analysis on novel

based open source network simulation tools for mo-

bile ad hoc networks. volume 5, pages 716–721.

Rouainia, H., Grichi, H., Kahloul, L., and Khalgui, M.

(2020). 3d mobility, resizing and mobile sink nodes

in reconﬁgurable wireless sensor networks based on

multi-agent architecture under energy harvesting con-

straints. In Proceedings of ICSOFT 2020, volume 97,

pages 394–406. ScitePress.

Rouainia, H., Grichi, H., Kahloul, L., and Khalgui, M.

(2022). New energy efﬁcient and fault tolerant

methodology based on a multi-agent architecture in

reconﬁgurable wireless sensor networks. pages 405–

417.

RWiN (2016). RWiN-Project: New Solutions for Reconﬁg-

urable Wireless Sensor Networks. https://lisi-lab.wix.

com/rwinproject. [Online; accessed 26-March-2022].

RWSNSim (2022). RWSNSim: Reconﬁgurable Wireless

Sensor Networks Simulator. https://hanenerouainia.

wixsite.com/rwsnsim. [Online; accessed 20-April-

2022].

Vijayalakshmi, S. and Muruganand, S. (2018). Wireless

Sensor Network: Architecture - Applications - Ad-

vancements. Mercury Learning & Information, 1 edi-

tion.

Reconﬁgurable Wireless Sensor Networks Simulator (RWSNSim): A New Discrete-event Simulator

361